按照tutorial来在cmd里面输入:input output 文件名和type即可。
data set在webgpu的页面有的下载。
F:\1MOOC_SourceCode\HPP\libwb\build\Debug>.\MP1.exe -e -i input2.raw,input1.raw -o out.raw -t vector
http://webgpu.hwu.crhc.illinois.edu/mp/1
Local Development
While not required, the library used throughout the course can be downloaded from Github. The library does not depend on any external library (and should be cross platform), you can use make to generate the shared object file (further instructions are found on the Github page). Linking against the library would allow you to get similar behavior to the web interface (minus the imposed limitations). Once linked against the library, you can launch the your program as follows:
./program -e <expected_output_file> -i <input_file_1>,<input_file_2> -o <output_file> -t <type>
The <expected_output_file> and <input_file_n> are the input and output files provided in the dataset. The <output_file> is the location you’d like to place the output from your program. The <type> is the output file type: vector, matrix, or image. If an MP does not expect an input or output, then pass none as the parameter.
In case of issues or suggestions with the library, please report them through the issue tracker on Github.
本文介绍如何使用WebGPU进行本地开发,包括如何设置环境、运行示例程序以及使用特定命令行参数来处理输入输出文件。提供了详细的步骤说明,并附上了程序运行实例。

被折叠的 条评论
为什么被折叠?



